AttributeError: '_UserObject' object has no attribute 'predict'

[madmax01]

hi, getting following Error when running

python3 main.py -A -odds=fanduel

Traceback (most recent call last): File "main.py", line 111, in main() File "main.py", line 100, in main NN_Runner.nn_runner(data, todays_games_uo, frame_ml, games, home_team_odds, away_team_odds) File "NBA-Machine-Learning-Sports-Betting\src\Predict\NN_Runner.py", line 18, in nn_runner ml_predictions_array.append(model.predict(np.array([row]))) AttributeError: '_UserObject' object has no attribute 'predict'

[Skysafe64]

python3 main.py -xgb -odds=fanduel

[madmax01]

this is just xgboost model Predictions.

tried to get "-A" working with including the Neural Network...

when doing

python3 main.py -nn -odds=fanduel (its same error).

was this option removed or so?

[nkgilley]

The option was not removed, it's working fine for me.

What repo/branch are you running?

Are you running in Google Colab or local?

[madmax01]

i used this locally "not google colab"

https://github.com/kyleskom/NBA-Machine-Learning-Sports-Betting.git

and tried then to run python3 main.py -A -odds=fanduel

so to have the xgboost and neural check.

anything i miss here? ;)

[nkgilley]

What commit are you on?

git rev-parse HEAD

[madmax01]

i just downloaded the zip and done. assuming zip is based on master? ;)

python -m Get_Data python -m Create_Games

Train models

cd ../Train-Models python -m XGBoost_Model_ML python -m XGBoost_Model_UO

and python3 main.py -A -odds=fanduel

[madmax01]

git rev-parse HEAD "0339624c68948c23e257f2105f51f88753d93b25"

[nkgilley]

out of date. do a git pull or download the latest zip

[madmax01]

i downloaded the latest zip.

how can i do git update after downloaded zip?

[madmax01]

just git pull. thats it?

[madmax01]

remote: Enumerating objects: 19, done. remote: Counting objects: 100% (19/19), done. remote: Compressing objects: 100% (10/10), done. remote: Total 19 (delta 9), reused 17 (delta 9), pack-reused 0 Unpacking objects: 100% (19/19), 1.71 MiB | 3.82 MiB/s, done. From https://github.com/kyleskom/NBA-Machine-Learning-Sports-Betting 0339624c..d5300ac1 master -> origin/master error: Your local changes to the following files would be overwritten by merge: src/Utils/tools.py Please commit your changes or stash them before you merge. Aborting Updating 0339624c..d5300ac1

it says 0339624c68948c23e257f2105f51f88753d93b25

[madmax01]

ah i done a stash and pull again now

d5300ac15b0f0ed0d3b861def9bafd99ba205403

seems updated 3h ago.

but still getting for NN_Runner.py following error

the AttributeError: '_UserObject' object has no attribute 'predict'

[madmax01]

may any hint what i need to fix here?

[JBmartin1991]

In NN_Runner, its pulling the models from "Models/NN_Models/Trained-Model-ML" and "Models/NN_Models/Trained-Model-OU", which hasnt been generated because you generated XGBoost data instead with the commands "python -m XGBoost_Model_ML" and "python -m XGBoost_Model_UO". However I dont see any way to generate the NN models in the code

[madmax01]

i did run now python -m and each below

src\Trained-Models\Train_model src\Trained-Models\Train_model_UO

but seems still same error... not finding an info in the Documentation how to deal with the keras

[madmax01]

tried to add bit debug

`import copy import numpy as np import pandas as pd import tensorflow as tf from colorama import Fore, Style, init, deinit from tensorflow.keras.models import load_model from src.Utils import Expected_Value

init()

try: model = load_model('Models/NN_Models/Trained-Model-ML') ou_model = load_model("Models/NN_Models/Trained-Model-OU") except OSError as e: print(f"Failed to load model file: {e}") raise SystemExit(1)

if not isinstance(model, tf.keras.models.Model): print("Loaded model is not a Keras model") raise SystemExit(1)

def nn_runner(data, todays_games_uo, frame_ml, games, home_team_odds, away_team_odds): ml_predictions_array = []

for row in data:
    ml_predictions_array.append(model.predict(np.array([row])))

frame_uo = copy.deepcopy(frame_ml)
frame_uo['OU'] = np.asarray(todays_games_uo)
data = frame_uo.values
data = data.astype(float)
data = tf.keras.utils.normalize(data, axis=1)

ou_predictions_array = []

for row in data:
    ou_predictions_array.append(ou_model.predict(np.array([row])))

count = 0
for game in games:
    home_team = game[0]
    away_team = game[1]
    winner = int(np.argmax(ml_predictions_array[count]))
    under_over = int(np.argmax(ou_predictions_array[count]))
    winner_confidence = ml_predictions_array[count]
    un_confidence = ou_predictions_array[count]
    if winner == 1:
        winner_confidence = round(winner_confidence[0][1] * 100, 1)
        if under_over == 0:
            un_confidence = round(ou_predictions_array[count][0][0] * 100, 1)
            print(Fore.GREEN + home_team + Style.RESET_ALL + Fore.CYAN + f" ({winner_confidence}%)" + Style.RESET_ALL + ' vs ' + Fore.RED + away_team + Style.RESET_ALL + ': ' +
                  Fore.MAGENTA + 'UNDER ' + Style.RESET_ALL + str(todays_games_uo[count]) + Style.RESET_ALL + Fore.CYAN + f" ({un_confidence}%)" + Style.RESET_ALL)
        else:
            un_confidence = round(ou_predictions_array[count][0][1] * 100, 1)
            print(Fore.GREEN + home_team + Style.RESET_ALL + Fore.CYAN + f" ({winner_confidence}%)" + Style.RESET_ALL + ' vs ' + Fore.RED + away_team + Style.RESET_ALL + ': ' +
                  Fore.BLUE + 'OVER ' + Style.RESET_ALL + str(todays_games_uo[count]) + Style.RESET_ALL + Fore.CYAN + f" ({un_confidence}%)" + Style.RESET_ALL)
    else:
        winner_confidence = round(winner_confidence[0][0] * 100, 1)
        if under_over == 0:
            un_confidence = round(ou_predictions_array[count][0][0] * 100, 1)
            print(Fore.RED + home_team + Style.RESET_ALL + ' vs ' + Fore.GREEN + away_team + Style.RESET_ALL + Fore.CYAN + f" ({winner_confidence}%)" + Style.RESET_ALL + ': ' +
                  Fore.MAGENTA + 'UNDER ' + Style.RESET_ALL + str(todays_games_uo[count]) + Style.RESET_ALL + Fore.CYAN + f" ({un_confidence}%)" + Style.RESET_ALL)
        else:
            un_confidence = round(ou_predictions_array[count][0][1] * 100, 1)
            print(Fore.RED + home_team + Style.RESET_ALL + ' vs ' + Fore.GREEN + away_team + Style.RESET_ALL + Fore.CYAN + f" ({winner_confidence}%)" + Style.RESET_ALL + ': ' +
                  Fore.BLUE + 'OVER ' + Style.RESET_ALL + str(todays_games_uo[count]) + Style.RESET_ALL + Fore.CYAN + f" ({un_confidence}%)" + Style.RESET_ALL)
    count += 1

print("--------------------Expected Value---------------------")
count = 0
for game in games:
    home_team = game[0]
    away_team = game[1]
    ev_home = float(Expected_Value.expected_value(ml_predictions_array[count][0][1], int(home_team_odds[count])))
    ev_away = float(Expected_Value.expected_value(ml_predictions_array[count][0][0], int(away_team_odds[count])))
    if ev_home > 0:
        print(home_team + ' EV: ' + Fore.GREEN + str(ev_home) + Style.RESET_ALL)
    else:
        print(home_team + ' EV: ' + Fore.RED + str(ev_home) + Style.RESET_ALL)

    if ev_away > 0:
        print(away_team + ' EV: ' + Fore.GREEN + str(ev_away) + Style.RESET_ALL)
    else:
        print(away_team + ' EV: ' + Fore.RED + str(ev_away) + Style.RESET_ALL)
    count += 1

deinit()`

if i do that > it tells "Loaded model is not a Keras model"

may the Loading process is not correct? xgboost load json but inside the NN folder are pb files

[Calle1410]

I also have the same error. I am on git rev-parse HEAD: b07e8c5150d2c9b0b34696a2436a9703ee2d32eb, when I try to do a git pull it says "Already up to date". I use python 3.10.10, tensorflow 2.11.0 and keras 2.11.0. I tried to fix it myself but got to a point where I dont get further. Maybe someone knows how to solve that issue.

NBA-Machine-Learning-Sports-Betting\src\Predict\NN_Runner.py", line 18, in nn_runner ml_predictions_array.append(model.predict(np.array([row]))) AttributeError: '_UserObject' object has no attribute 'predict'

[lachgil]

I am also having issues with the latest version, AttributeError: '_UserObject' object has no attribute 'predict' I tried the second suggestion with no luck; https://stackoverflow.com/questions/68173923/error-userobject-object-has-no-attribute-predict I am running this locally with head: b07e8c5150d2c9b0b34696a2436a9703ee2d32eb Python 3.10.10 TF: 2.11.0 keras: 2.11.0

[kyleskom]

Try python 3.9

[Calle1410]

Sadly python 3.9 doesn´t solve the issue. Same problem as with 3.10

[nkgilley]

looks like this is happening with windows users maybe? does anybody using windows have it working?

[marine1983boots]

Ive got this working on my windows machine with Pycharm. Take a look at the discussion tabs

on NN_Runner.py

Change

from tensorflow.keras.models import load_model

to

from tensorflow.python.keras.models import load_model

Also check if you're a UK user. Try the bookie as bet365 instead of fanduel.

[XiangjunWu]

Also had the issue: AttributeError: '_UserObject' object has no attribute 'predict'

I was able to solve it by training the model for Train_Model:

Train models

cd ../Train-Models python -m Train_Model python -m Train_Model_UO

Due to some reason, the file was not saved correctly, so I had to move the files from "Trained-Model-ML-1679137211.9821796" to the folder "Trained-Model-ML" and the same for "Trained-Model-OU-1679133914.0189226" to "Trained-Model-OU".

Then the error message disappears.

[lachgil]

I ran both , python -m Train_Model & Train_Model_UO, and after moving over the folders into NN_Models this is working correctly for me on windows. Thanks!